Challenges
Datasets
Workspaces
Papers
Discussions
Leaderboard
Log inSign up
Challenges
Datasets
Workspaces
Papers
Discussions
Leaderboard
Blog
Job Board
Q3AS

© 2026 Aqora Quantum S.A.S.

TermsPrivacyLegal Notice

Lecture notes in computer science

Non-Interactive Distributed Point Functions

Elette Boyle, Lalita Devadas, Sacha Servan-Schreiber

Published Jan 1, 2025• 6 citations
Publisher
DOI
10.1007/978-3-031-91820-9_1

Related papers

Programmable Distributed Point Functions

Elette Boyle, Niv Gilboa, Yuval Ishai +1 more

Lecture notes in computer science•Jan 1, 2022•20 citations

Criticality in Neural Network Function Space through Wilsonian Fixed Points and Finite-Width Corrections

Eric Howard, Iftekher S. Chowdhury, Hardique Dasore +1 more

Neural networks can be studied not only as parameterized computational models but also as probability distributions over functions. In this paper we develop a Wilsonian interpretation of criticality in the neural network-quantum field theory correspondence, treating the infinite width Gaussian-process limit as a free field fixed point and finite width corrections as perturbations that create non-Gaussian interactions. In this way, we do not see the departure from infinite width as a small approximation error but as the process by which interaction, complexity, expressivity and phase-like behaviour enter neural network function space. Width, depth, activation nonlinearity, initialization variance, and training dynamics are considered as control parameters that change the effective action of the network ensemble. The critical regime is when the higher-order connected correlation functions become non-negligible, and when the finite width operators become relevant or marginal scaling factors, and when the function distribution becomes sensitive to scale-dependent structure. There we can view overparameterization as suppressing a relationship of interacting terms and we find that the critical structure of finite neural networks is given by finite-width effects. This framework is a theoretical basis for studying trainability, generalization, and architectural universality of neural network functions through Wilsonian fixed points, perturbations, and critical surfaces.

cond-mat.dis-nnCross-listedarXiv:2608.20403•Aug 5, 2026

Multi-party Distributed Point Functions, Revisited

Elaine Shi, 顾天瑶, Xuanye Zheng +3 more

Lecture notes in computer science•Nov 5, 2026•0 citations

Long-Distance Behaviors of the Pair Distribution Function of Degenerate and Non-Degenerate Bose Fluids

Tohru Morita, Hiroaki Hara

The formalism proposed for degenerate Bose fluids by one of the present authors [T. Morita, Prog. Theor. Phys. 39 (1968), 1100J is extended for the calculation of the one- and two-particle distribution functions. With the aid of the diagram expansion, it is shown that the long-distance l/r and 1/r2 behaviors of the, pair distribution function of the degenerate ideal Bose are exactly canceled in degenerate interacting Bose fluids in general. In the proof, it is assumed that vk/k->= as k-70 for the weakly interacting case and.sa(k, 0) /k-0= as k-00 for the general case, where Vk and .saCk, 0) are the Fourier transforms of interacting potential and of a generalized binary collision kernel, respectively. It is stressed that the Hartree-Fock approximation is insufficient in showing the cancellation of the l/r2 term. In fact, the cancellation is attained by taking account of the averages of the type: <afo-I-k ak ap) and <ap* ak* ap-I,k). For non-degenerate case, the 1/7-2 behavior, which appears Just above the A-point of the ideal Bose gas is als9 shown to be canceled in interacting Bose fluids.

Progress of Theoretical Physics•Jan 1, 1969•5 citations

Many-particle correlations in a non-equilibrium Luttinger liquid

I V Protopopov, D B Gutman, A. D. Mirlin

We develop an operator-based approach to the problem of Luttinger liquid conductor in a non-equilibrium stationary state. We show that the coherent-state many-body fermionic density matrix as well as all fermionic correlation functions out of equilibrium are given by one-dimensional functional determinants of the Fredholm type. Thus, the model constitutes a remarkable example of a many-body problem where all the correlation functions can be evaluated exactly. On the basis of the general formalism we investigate four-point correlation functions of the fermions coming out of the Luttinger liquid wire. Obtained correlations in the fermionic distribution functions represent the combined effect of interaction and non-equilibrium conditions.

arXiv:1107.5561•Journal of Statistical Mechanics Theory and Experiment•Nov 2, 2011•16 citations

Point Interactions

IMPERIAL COLLEGE PRESS eBooks•Mar 1, 2006•0 citations
Neural networks can be studied not only as parameterized computational models but also as probability distributions over functions. In this paper we develop a Wilsonian interpretation of criticality in the neural network-quantum field theory correspondence, treating the infinite width Gaussian-process limit as a free field fixed point and finite width corrections as perturbations that create non-Gaussian interactions. In this way, we do not see the departure from infinite width as a small approximation error but as the process by which interaction, complexity, expressivity and phase-like behaviour enter neural network function space. Width, depth, activation nonlinearity, initialization variance, and training dynamics are considered as control parameters that change the effective action of the network ensemble. The critical regime is when the higher-order connected correlation functions become non-negligible, and when the finite width operators become relevant or marginal scaling factors, and when the function distribution becomes sensitive to scale-dependent structure. There we can view overparameterization as suppressing a relationship of interacting terms and we find that the critical structure of finite neural networks is given by finite-width effects. This framework is a theoretical basis for studying trainability, generalization, and architectural universality of neural network functions through Wilsonian fixed points, perturbations, and critical surfaces.
The formalism proposed for degenerate Bose fluids by one of the present authors [T. Morita, Prog. Theor. Phys. 39 (1968), 1100J is extended for the calculation of the one- and two-particle distribution functions. With the aid of the diagram expansion, it is shown that the long-distance l/r and 1/r2 behaviors of the, pair distribution function of the degenerate ideal Bose are exactly canceled in degenerate interacting Bose fluids in general. In the proof, it is assumed that vk/k->= as k-70 for the weakly interacting case and.sa(k, 0) /k-0= as k-00 for the general case, where Vk and .saCk, 0) are the Fourier transforms of interacting potential and of a generalized binary collision kernel, respectively. It is stressed that the Hartree-Fock approximation is insufficient in showing the cancellation of the l/r2 term. In fact, the cancellation is attained by taking account of the averages of the type: <afo-I-k ak ap) and <ap* ak* ap-I,k). For non-degenerate case, the 1/7-2 behavior, which appears Just above the A-point of the ideal Bose gas is als9 shown to be canceled in interacting Bose fluids.
We develop an operator-based approach to the problem of Luttinger liquid conductor in a non-equilibrium stationary state. We show that the coherent-state many-body fermionic density matrix as well as all fermionic correlation functions out of equilibrium are given by one-dimensional functional determinants of the Fredholm type. Thus, the model constitutes a remarkable example of a many-body problem where all the correlation functions can be evaluated exactly. On the basis of the general formalism we investigate four-point correlation functions of the fermions coming out of the Luttinger liquid wire. Obtained correlations in the fermionic distribution functions represent the combined effect of interaction and non-equilibrium conditions.